Re: [Taverna-hackers] Proposal - fold reference translate functionality into DataManager

[Stian S. <sso...@cs...>]

On Jan 29, 2008 4:41 PM, Tom Oinn <mer...@ca...> wrote:

> I propose we add a method to DataManager which is the same as the
> current resolver but with the ability to specify a set of reference
> scheme types which must be (all or at least one depending on semantics
> of the call) present in the returned data document, or ignored if
> there's a collection. This would be an additional getEntity method with
> the extra information. The translator infrastructure then hangs on the
> side of DataManager without ever having to be directly exposed to the
> client code.

My worry is that the DataManager ends up very thick, it is both
responsible for storing things (in memory, files, database), and
additionally resolve things in complex ways, such as by using
translation or requesting from other peers.

I believe we perhaps need to split into a DataManagerStore (doing the
boring backend stuff) and a DataManager(Front) to avoid the growing
AbstractDataManager. In that picture the DataManagerStore is "hidden"
from most clients, and can have some strict update-mechanism, usable
both for translations and incoming things from peers. (Maybe the peers
want to announce that they've done a translation, for instance)

-- 
Stian Soiland, myGrid team
School of Computer Science
The University of Manchester
http://www.cs.man.ac.uk/~ssoiland/